Professor Yu-Chi Wu received both his M.S. and Ph.D. degrees in Electrical and Computer Engineering from the Georgia Institute of Technology (Georgia Tech), U.S.A. in 1993. Since 1986, he has been actively engaged in both academic and industrial pursuits. In 1994, he joined the Lien-Ho Institute of Technology and Commerce (later renamed National United University), Taiwan, where he has served in various leadership roles, including Chairman of the Department of Electrical Engineering (1995-1998), Director of the Library (2000-2001 and 2016-2018), Dean of Student Affairs (2003-2006), Dean of the College of Electrical Engineering and Computer Science (2012-2015), and Vice President (2016-2017). He was a visiting scholar at the Alexandria Research Institute, Virginia Polytechnic Institute and State University, U.S.A. (November 2002-March 2003), the University of New South Wales, Australia (July-November 2010), and the University of Texas at Austin, USA (July-August and December 2016). He served as the Chair of PE-31, IEEE Taipei Chapter, from 2022-2025. He is currently an IEEE Senior Member.
Professor Wu's research interests encompass AI applications in Remaining Useful Life (RUL) estimation of power supplies and mold anomaly detection, smart manufacturing, IMU-based gait measurement and motion capture, wearable devices, automated guided vehicles, power system resilience, operations of power systems and microgrids, optimization algorithms, wind generator design, Pulley motor design, mobile health management, and energy-saving technologies. He has published over 100 papers in these fields.
His outstanding contributions have been recognized through numerous awards, including Class A Research Awards from the National Science Council of Taiwan, Class A Research Awards from the Lien-Ho Education Foundation of Technology and Commerce, Distinguished Alumni Awards from the National Kaohsiung University of Applied Sciences, First Prize in the Energy Saving Contest of Schneider Electrical Engineering Cup, Outstanding Specialist Awards from the National Science Council of Taiwan, Gold and Bronze Medals at the Taipei International Invention Show and Technomart, Bronze Medals at Le Concours Lepine, and Best Paper Awards at various international conferences.
Smart System Lab is a lab run by several professors at National United University who work across several departments (electrical engineering, computer science and information engineering, and information management). The projects conducted by this Lab are related to machine learning, wearable devices for healthcare, smart grids, smart manufacturing, and intelligent power system operations.
remaining useful life for switch-mode power supplies, machine learning for IMU denoising and error reduction, Wearable devices for motion capture, power system resilience, optimization algorithms, Mobile health management, IOT, and smart manufacturing.
He has published over 100 papers in various areas. He has also won several research and invention awards: Class A Research Awards of National Science Council of Taiwan, Class A Research Awards of Lien-Ho Education Foundation of Technology and Commerce, Distinguished Alumni Awards of National Kaohsiung University of Applied Sciences, First Prize of Energy Saving Contest of Schneider Electrical Engineering Cup, Outstanding Specialist Awards of National Science Council of Taiwan, Gold Medal and Bronze Medals of Taipei International Invention Show and Technomart, Bronze Medals of Le Concours Lepine, and Best Paper Awards of various international conferences. Professor Wu is a senior member of IEEE and was the chairman of PE-31 of the IEEE Taipei Chapter from 2022 to 2025.
Professor Yu-Chi Wu received both his M.S. and Ph.D. degrees in Electrical and Computer Engineering from the Georgia Institute of Technology (Georgia Tech), U.S.A. in 1993.
Job Description
Monte Carlo analysis (MC analysis) is an essential tool for assessing power system resilience. It can help identify critical events that may lead to system vulnerabilities or failures, and, therefore, is crucial for prioritizing resilience measures and developing strategies to mitigate the impact of specific events. However, from a computational perspective, MC analysis is quite challenging. Due to the large number of interconnected components and various uncertainties in power systems, the input space for MC analysis can be high-dimensional. This makes it difficult to obtain statistically meaningful results and increases computational requirements. When Monte Carlo analysis is conducted alongside Optimal Power Flow (OPF) analysis, the above challenges are further compounded. The OPF analysis, which calculates the load loss for each MC time step, can derive Key Performance Indicators (KPIs), such as ENS (energy not served) and survivability. Therefore, MC analysis combined with OPF (MC-OPF) is computationally intensive and challenging. This project aims to find an efficient and suitable AI model for MC-OPF to reduce the computational load and time. The AI model is trained on a few MC scenarios and OPF solutions and effectively provides system-resilience analysis results.
Preferred Intern Educational Level
Graduate student or senior undergraduate student
Skill sets or Qualities
Python, Matlab/SimScape, AI/machine learning
Job Description
A single-phase AC-DC switch-mode power supply (AC-DC SMPS) offers lower transportation costs and variable power conversion characteristics, making it widely used in industrial machinery, tools, and control systems worldwide. However, if the SMPS fails, it may lead to control errors or failures, causing losses and safety hazards. Therefore, it is necessary to study the prediction of SMPS performance degradation and remaining useful life (RUL). The main task is to develop an accurate physical model of the SMPS using a simulator. This model incorporates the degradation characteristics of key components, enabling rapid simulation of degradation dynamics and the determination of SMPS degradation characteristics. The actual degradation data of SMPS are then measured in a Highly Accelerated Life Testing (HALT) environment. An estimation method based on machine learning models will be developed to estimate RUL. The research results can effectively predict the remaining useful life of single-phase AC-DC power supplies widely used in industrial systems.
Preferred Intern Educational Level
Graduate student or senior undergraduate student
Skill sets or Qualities
Python, Matlab/SimScape, AI/Machine Learning